Switch to DuckDuckGo Search
   May 19, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >

Toggle Join/Part | bottom
[00:30:41] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[00:31:19] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has joined #illumos
[01:35:54] *** BrownBear <BrownBear!~BrownBear@107.175.247.247> has joined #illumos
[02:44:39] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has joined #illumos
[03:11:54] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[03:12:49] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[03:16:55] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[03:45:27] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[04:12:38] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has joined #illumos
[04:12:39] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[04:13:00] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[04:16:22] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has quit IRC (Ping timeout: 272 seconds)
[04:21:31] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (Ping timeout: 260 seconds)
[04:22:35] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[06:05:51] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (Ping timeout: 265 seconds)
[06:08:23] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[06:10:21] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[06:11:10] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[06:20:46] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has quit IRC (Read error: Connection reset by peer)
[06:39:47] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[06:40:20] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[06:41:49] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[06:42:30] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[07:10:16] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[07:10:41] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[07:12:00] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[07:12:51] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[07:28:11] *** kovert <kovert!~kovert@204.141.173.249> has quit IRC (Ping timeout: 260 seconds)
[07:29:02] *** kovert <kovert!~kovert@204.141.173.249> has joined #illumos
[07:31:38] *** jemershaw <jemershaw!~jemershaw@c-68-83-252-28.hsd1.pa.comcast.net> has quit IRC (Remote host closed the connection)
[07:34:03] *** jemershaw <jemershaw!~jemershaw@c-68-83-252-28.hsd1.pa.comcast.net> has joined #illumos
[07:40:25] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[07:41:01] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[07:54:29] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a00:eb45:9acb:d434:5802:3a63> has joined #illumos
[08:25:26] *** pwinder <pwinder!~pwinder@86.11.191.8> has joined #illumos
[08:30:48] *** phyre <phyre!~phyre___@78.30.22.107> has joined #illumos
[08:32:05] *** fanta1 <fanta1!~fanta1@p200300f76bc46200ace3256a4f32ea36.dip0.t-ipconnect.de> has joined #illumos
[08:34:48] *** BOKALDO <BOKALDO!~BOKALDO@87.110.101.147> has joined #illumos
[08:37:15] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Remote host closed the connection)
[08:37:42] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[08:39:00] *** paulf <paulf!020771b0@lfbn-lyo-1-274-176.w2-7.abo.wanadoo.fr> has joined #illumos
[08:40:52] <paulf> Does anyone know if any work on maintaining Valgrind has been done in the Ullumos community?
[08:40:52] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[08:41:12] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[08:43:03] <paulf> I haven't tested on any OpenSolaris derived builds, but I tried Valgrind 3.16 RC1 the other day on Solaris 11.4 and there were build issues and many regression test failures
[08:43:03] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[08:43:22] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[09:02:29] *** paulf <paulf!020771b0@lfbn-lyo-1-274-176.w2-7.abo.wanadoo.fr> has quit IRC (Remote host closed the connection)
[09:02:40] <rzezeski> paulf: There is a valgrind package listed in OpenIndiana, version 3.15, so that's a good sign. Other than that I'm not sure how well it works or who maintains it. If you are just trying to verify memory usage you could also look at libumem, which is the more native way of doing such things.
[09:10:10] *** jimklimov1 <jimklimov1!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has quit IRC (Ping timeout: 256 seconds)
[09:10:10] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[09:10:33] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[09:11:59] *** andy_js <andy_js!~andy@94.2.168.219> has joined #illumos
[09:12:00] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[09:12:44] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[09:13:03] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[09:13:55] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[09:14:34] *** neuroserve <neuroserve!~toens@ip-84-118-128-44.unity-media.net> has quit IRC (Ping timeout: 240 seconds)
[09:18:52] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[09:25:17] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-ucmzivxneadmfnyn> has quit IRC (Ping timeout: 252 seconds)
[09:27:29] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-zkyzxetsulvvxvmg> has joined #illumos
[09:27:50] *** neuroserve <neuroserve!~toens@ip-94-114-252-163.unity-media.net> has joined #illumos
[09:30:01] *** paulf <paulf!c05e1f02@nat-ies.mentorg.com> has joined #illumos
[09:35:09] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has joined #illumos
[09:47:41] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has joined #illumos
[10:12:13] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[10:13:07] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[10:13:58] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[10:14:18] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[10:18:44] *** kohju <kohju!~kohju@gw.justplayer.com> has quit IRC (Excess Flood)
[10:19:11] *** kohju <kohju!~kohju@gw.justplayer.com> has joined #illumos
[10:29:24] <jimklimov> Hi all, FWIW from my daily inbox straight to you: Jim Salter published a nicely written benchmarking research into various setups of ZFS over 8 disks, vs mdadm+ext4: https://arstechnica.com/gadgets/2020/05/zfs-versus-raid-eight-ironwolf-disks-two-filesystems-one-winner/
[10:40:50] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[10:41:28] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[10:41:46] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[10:42:38] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[10:43:14] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[10:43:48] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[10:53:13] <igork> jimklimov: thanks for info
[11:08:31] <jollyd> paulf, rzezeski: Ivo Raisr and Petr Pavlu ported Valgrind to Solaris. I originally added their port (hosted on bitbucket) to oi-userland for testing before it was integrated, I think Michal Nowak bumped to the latest version some months ago.
[11:11:32] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Read error: Connection reset by peer)
[11:11:58] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has joined #illumos
[11:25:10] <paulf> I know about the Solaris port history. Have there been any fixes recently? I may be able to get any changes merged upstream.
[11:27:52] *** am11 <am11!~am11@dsl-hkibng22-54f985-221.dhcp.inet.fi> has quit IRC (Quit: WeeChat 1.9.1)
[11:29:13] <paulf> looking at the oi-userand github, it seems not
[11:29:15] <jollyd> paulf: the build recipe is located at https://github.com/OpenIndiana/oi-userland/tree/oi/hipster/components/developer/valgrind , there does not seem to be any outstanding patch currently, but I remember that illumos support is incomplete due to differences with Solaris
[11:30:35] <jollyd> a year ago already... seemed like a few months.. time flies
[11:31:49] <jollyd> paulf: I can look at 3.16 but not before mid-June.
[11:33:37] <jollyd> paulf: do you know when the 3.16 release would happen?
[11:34:32] <paulf> OK, atm I don't have any illumos system of any sort, just one Solaris 11.4 VM. There was one build patch that I submitted a few weeks ago which has been integrated.
[11:35:46] <paulf> Otherwise there is one other build issue on Solaris related to using -std=gnu99 and one runtime issue that seems to be related to dtrace data in the debuginfo elf sections.
[11:36:08] <paulf> 3.16 just reached RC1 yesterday
[11:37:11] <jollyd> paulf: OK, I can start a build right now, I have a few minutes (coffee break)
[11:38:12] <paulf> There may be other runtime issues - the debuginfo problem is polluting over 90% if the regression tests on Solaris
[11:38:16] <paulf> OK thanks
[11:40:09] *** kohju <kohju!~kohju@gw.justplayer.com> has quit IRC (Excess Flood)
[11:40:09] <jollyd> paulf: do you know where I can fetch a tarball of the rc1?
[11:40:32] *** kohju <kohju!~kohju@gw.justplayer.com> has joined #illumos
[11:40:33] <paulf> If you have the same build problem that I saw you will need to run configure CFLAGS="-std=gnu99"
[11:41:46] <paulf> I usually just do a git clone of https://sourceware.org/git/valgrind.git
[11:50:56] <jollyd> paulf: building now
[11:51:37] <jollyd> paulf: which compiler do you use? the default in oi-userland is gcc 7.5.0.
[11:51:56] <paulf> gcc should be fine
[11:52:25] <paulf> but you probably need to use GNU make
[11:53:09] <jollyd> paulf: yes it is the case in oi-userland, we use GNU make by default
[11:54:43] <jollyd> paulf: we do pass -std=gnu99 -D__EXTENSIONS__ I commented "Flags are inferred from CC and tests rely on sigaction" should date back to 3.11
[11:59:54] <jollyd> paulf: usr/lib/$(MACH64)/valgrind/exp-sgcheck-amd64-solaris is not provided anymore, is it intended?
[12:00:28] <paulf> Yes. exp-sgcheck was removed recently.
[12:03:18] <jollyd> "NEW PACKAGE CONTENTS ARE LOCALLY VALIDATED AND READY TO GO", now time to test
[12:07:23] *** Teknix <Teknix!~pds@69.41.134.110> has quit IRC (Ping timeout: 265 seconds)
[12:09:41] *** Teknix <Teknix!~pds@69.41.134.110> has joined #illumos
[12:14:30] <jollyd> paulf: I see a number of failed tests, I will upload the results after the testsuite has finished.
[12:36:05] <jollyd> paulf: https://github.com/OpenIndiana/oi-userland/pull/5765 with test results
[12:36:56] <jollyd> 783 tests, 70 stderr failures, 18 stdout failures, 3 stderrB failures, 6 stdoutB failures, 7 post failures
[12:40:22] <jollyd> paulf: let me know if it is consistent with your findings on Solaris
[12:43:51] <paulf> That's much better than my Solaris results. All but about 20 were failing, so I assume oi hasn't had the change to dtrace_date that impacts Solaris.
[12:44:09] <paulf> dtrace_data
[12:55:18] <jollyd> paulf: for reference you can find the tarball with test results at http://dlc-int.openindiana.org/users/aurelien/testing/valgrind-3.16rc1-tests.tar.gz
[12:58:15] <jollyd> it seems there are a few false negative due to difference in expected output
[13:00:38] <jollyd> paulf: need to go back to real work but let me know how I can help further.
[13:11:22] <paulf> I mainly wanted to know the status, so thanks for the update. It seems like things are reasonably functional on oi.
[13:34:17] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a00:eb45:9acb:d434:5802:3a63> has quit IRC (Ping timeout: 260 seconds)
[13:53:26] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has joined #illumos
[13:57:22] *** fanta1 <fanta1!~fanta1@p200300f76bc46200ace3256a4f32ea36.dip0.t-ipconnect.de> has quit IRC (Quit: fanta1)
[14:26:01] <sjorge> jbk did you have a chance to peak at the ldapclient/smbserver issue/feature I filed?
[14:26:11] <sjorge> My work around is sort of working OK, but it's rather hackish
[14:35:08] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:47:59] *** phyre <phyre!~phyre___@78.30.22.107> has quit IRC (Remote host closed the connection)
[14:51:30] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[15:19:29] *** fanta1 <fanta1!~fanta1@p200300f76bc46200ace3256a4f32ea36.dip0.t-ipconnect.de> has joined #illumos
[15:36:01] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[15:36:38] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has joined #illumos
[15:51:13] *** KungFuJesus <KungFuJesus!~adam@72.49.160.91> has quit IRC (Ping timeout: 264 seconds)
[16:13:59] <andyf> Can anyone tell me where uncompressed gets filled in before the call to lz4_compress here please?
[16:14:00] <andyf> http://src.illumos.org/source/xref/illumos-gate/usr/src/cmd/vtfontcvt/vtfontcvt.c#770
[16:15:51] <rmustacc> andyf: So it looks like write_glphys does it.
[16:16:06] <rmustacc> The thing tis that write_glyph_buf is a cb that writes to the global uncompressed.
[16:16:08] <andyf> ah yes
[16:16:27] <andyf> vtfontcvt is crashing in the lz4_compress() call on one of my systems
[16:16:30] <rmustacc> Though at first I overlooked it due to the & on write_glyph_bus.
[16:16:36] <andyf> (not on any other system I have access to though)
[16:16:44] <rmustacc> Interesting.
[16:17:11] <andyf> More interesting is that it does not crash with UMEM_DEBUG=default..
[16:17:21] <andyf> (it is linked with libumem)
[16:17:33] <rmustacc> The question that comes to mind is what are the contents of the tailq.
[16:22:07] *** fanta1 <fanta1!~fanta1@p200300f76bc46200ace3256a4f32ea36.dip0.t-ipconnect.de> has quit IRC (Quit: fanta1)
[16:25:27] <andyf> Everything looks good there. The buffer's constructed and the size is the same as the earlier calculation and malloc
[16:25:39] <andyf> but LZ4 runs off the end of memory somewhere
[16:26:27] <rmustacc> Hmm. We'll need to debug I guess. the fact that UMEM_DEBUG chases it away smells like uninitialized memory to me, tbh.
[16:27:05] <andyf> usr/src/common/lz4/lz4.c:884
[16:27:10] <andyf> if (A32(ref) == A32(ip)) {
[16:27:35] <rmustacc> Do the isize/osize args make sense?
[16:27:46] <andyf> ref is the bad pointer
[16:30:42] <andyf> lz4_compress:entry 89d2008,89d5008,1728,1728
[16:30:46] <andyf> looks fine
[16:33:22] <rmustacc> Hmm.
[16:34:27] <rmustacc> That code is a bit gnarly. I don't think I'm going to come to something quickly.
[16:34:49] <andyf> Same here - I'll get back to it later. The LZ4 code is not particularly transparent to me
[16:35:40] <andyf> Thanks.. my best guess so far is it's in the lz4 object (or the way that object is being generated on this particular system)
[16:43:01] <andyf> Well, not that.. a binary from a different system crashes too. I'll dig into it more later.
[17:00:12] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Quit: amrfrsh)
[17:06:12] <gitomat> [illumos-gate] 12693 Enable Forward Error Correction (FEC) configuration via dladm -- Paul Winder <paul at winder dot uk.net>
[17:13:31] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[17:16:44] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Remote host closed the connection)
[17:21:51] *** wacki <wacki!~wacki@i577A5F26.versanet.de> has joined #illumos
[17:37:58] *** jellydonut <jellydonut!~quassel@45.83.220.160> has quit IRC (Quit: jellydonut)
[17:39:37] *** jellydonut <jellydonut!~quassel@45.83.220.175> has joined #illumos
[17:52:01] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[17:57:32] <gitomat> [illumos-gate] 12679 want viona driver for bhyve -- Patrick Mooney <pmooney at pfmooney dot com>
[18:00:37] <jollyd> \o/
[18:01:04] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has quit IRC (Remote host closed the connection)
[18:05:12] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has joined #illumos
[18:06:37] <gitomat> [illumos-gate] 12738 Panic in vlapic_callout_handler -- Patrick Mooney <pmooney at pfmooney dot com>
[18:12:06] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[18:16:21] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[18:27:20] *** paulf <paulf!c05e1f02@nat-ies.mentorg.com> has quit IRC (Remote host closed the connection)
[18:49:06] <sjorge> Yay viona
[18:54:35] *** patdk-lap <patdk-lap!~patrickdk@208.94.189.75> has quit IRC (Remote host closed the connection)
[19:08:17] <andyf> To the tune of...?
[19:23:26] <jollyd> andyf: "My Sharona"?
[19:23:44] <jollyd> too doo doom doom "Yay Vi-o-na"
[19:24:45] <andyf> jollyd: that's the one I was thinking of :)
[19:24:51] <jollyd> :D
[19:32:34] *** pwinder <pwinder!~pwinder@86.11.191.8> has quit IRC (Quit: This computer has gone to sleep)
[19:33:16] *** ngchk1 <ngchk1!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has joined #illumos
[19:34:30] *** ngchk1_ <ngchk1_!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has quit IRC (Ping timeout: 258 seconds)
[19:44:02] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has quit IRC (Ping timeout: 256 seconds)
[19:45:24] <pmooney> any thoughts about how we should handle review for 12680 and 12682?
[19:45:52] <pmooney> their individual pieces were reviewed while going into illumos-joyent, but there aren't any reviewers which saw _all_ of it
[19:48:34] <pmooney> They are the last two consolidated commits with that problem
[19:48:52] <pmooney> the rest of the sync-up is going in piecemeal
[19:55:15] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[20:17:39] *** pwinder <pwinder!~pwinder@86.11.191.8> has joined #illumos
[20:21:15] <jlevon> pmooney: I can cover 12682 if you want to add me
[20:21:28] <pmooney> jlevon: that would be great, thanks
[20:21:35] <pmooney> I'm going 12680 myself right now
[20:22:17] <pmooney> blast radius is a little smaller on PPT
[20:23:29] *** pwinder <pwinder!~pwinder@86.11.191.8> has quit IRC (Quit: This computer has gone to sleep)
[20:36:12] *** arnoldoree <arnoldoree!~arnoldore@113.210.184.35> has quit IRC (Quit: Leaving)
[20:53:09] *** cantstanya <cantstanya!~chatting@gateway/tor-sasl/cantstanya> has quit IRC (Remote host closed the connection)
[21:10:07] <pmooney> jlevon: on the topic of a priv for hypervisor operations, what about one for pci pass-thru as well?
[21:10:32] <pmooney> we definitely still have an 'XXX' in ppt_open() suggesting that as a future enhancement
[21:19:34] <jlevon> pmooney: haven't really thought that one through...
[21:19:57] <pmooney> the consumer would obviously need access to the device as well
[21:20:29] <pmooney> but considering how there's room to make terrible mistakes in ppt, it might be the kind of thing that's better to keep off limits except for those specifically granted access
[21:21:06] <pmooney> I definitely remember the original ppt stuff from freebsd basically giving the user access to map any physical memory, due to its lack of checks for BAR mappings
[21:22:02] *** patdk-lap <patdk-lap!~patrickdk@208.94.189.75> has joined #illumos
[21:22:27] <jlevon> I can definitely see it making sense to keep separate from just hypervisor privs
[21:23:01] <pmooney> yeah, I would definitely keep it unloaded unless I had a need to use it
[21:23:51] <jlevon> I presume you're not thinking of how to allow specific devices
[21:23:53] <jlevon> just a blanket thing
[21:24:03] <pmooney> correct
[21:24:08] *** phyre <phyre!~phyre___@78.30.22.107> has joined #illumos
[21:24:28] <pmooney> I guess we can just guard it behind the hvm priv initially
[21:24:31] <pmooney> since it's useless w/o that
[21:24:38] * pmooney shrugs
[21:24:56] <jlevon> something that can come as a follow up for sure
[21:25:00] <pmooney> indeed
[21:28:54] <rmustacc> I think it would be interesting to do something like the more fine-grained privs that happened in S11 where you can do things like specify particular ports to the privlieged ports or files for FILE_DAC_READ in conjunction with pass throug,e tc.
[21:30:18] <jlevon> kind of a project tho
[21:33:54] <rmustacc> Yeah, not saying it should be done or block anything.
[21:44:27] *** neuroserve <neuroserve!~toens@ip-94-114-252-163.unity-media.net> has quit IRC (Ping timeout: 258 seconds)
[21:45:35] *** BOKALDO <BOKALDO!~BOKALDO@87.110.101.147> has quit IRC (Quit: Leaving)
[21:47:09] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has joined #illumos
[21:49:48] *** neuroserve <neuroserve!~toens@ip-94-114-236-17.unity-media.net> has joined #illumos
[21:54:09] *** teutat3s[m]1 <teutat3s[m]1!~teutat3sp@85.88.23.162> has joined #illumos
[22:06:27] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[22:35:36] <sjorge> pmooney: from looking at the ppt stuff with papertigers i think we do a lot less than what freebsd is doing in prepping the devices, still works though
[22:36:04] <pmooney> what are they doing that we aren't?
[22:36:39] <sjorge> memory hazy as I could barely read the code... but I think some resetting of the device?
[22:37:09] <sjorge> I think we pass it the guest does whatever and we get it back, repeat
[22:37:47] <rmustacc> We definitely should reset the device, at least the slot or do an FLR between things where it makes sense. Though for multi-funciton devices it's a bit more complicated.
[22:39:14] <sjorge> For nice and a hba it seems to be OK and working currently
[22:39:36] <sjorge> *nics
[22:41:34] *** neirac <neirac!~cneira@pc-4-149-45-190.cm.vtr.net> has joined #illumos
[22:49:23] *** nbhauke <nbhauke!~hauke@55d45c4f.access.ecotel.net> has quit IRC (Quit: nbhauke)
[22:57:02] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has joined #illumos
[22:59:58] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has quit IRC (Ping timeout: 260 seconds)
[23:03:00] *** jessfraz_ <jessfraz_!~jessfraz@unaffiliated/jessfraz> has quit IRC (Remote host closed the connection)
[23:04:43] *** jessfraz_ <jessfraz_!~jessfraz@unaffiliated/jessfraz> has joined #illumos
[23:08:42] *** nbhauke <nbhauke!~hauke@55d45c4f.access.ecotel.net> has joined #illumos
[23:09:25] *** wacki <wacki!~wacki@i577A5F26.versanet.de> has quit IRC (Quit: Lingo: www.lingoirc.com)
[23:10:57] *** jollyd <jollyd!~alarcher@2a01:cb1d:611:bb00:8820:7cbd:8965:af56> has quit IRC (Ping timeout: 272 seconds)
[23:11:59] *** jollyd <jollyd!~alarcher@2a01:cb1d:611:bb00:8820:7cbd:8965:af56> has joined #illumos
top

   May 19, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >